Cyclohexylphenyl glycolic acid (also referred to herein as "CHPGA") is used as a starting material for manufacturing compounds that have important biological and therapeutic activities. Such compounds include, for example, oxphencyclimine, oxyphenonium bromide, oxypyrronium bromide, oxysonium iodide, oxybutynin (4-diethylamino-2-butynyl phenylcyclohexylglycolate) and its metabolites, such as desethyloxybutynin (4-ethylamino-2-butynyl phenylcyclohexylglycolate). ##STR1##
The important relation between stereochemistry and biological activity is well known. For example, the (S)-enantiomers of oxybutynin and desethyloxybutynin have been shown to provide a superior therapy in treating urinary incontinence, as disclosed in U.S. Pat. Nos. 5,532,278 and 5,677,346. The (R) enantiomer of oxybutynin has also been suggested to be a useful drug candidate. [Noronha-Blob et al., J. Pharmacol. Exp. Ther. 256, 562-567 (1991)].
Racemic CHPGA is generally prepared by one of two methods: (1) selective hydrogenation of phenyl mandelic acid or of phenyl mandelate esters, as shown in Scheme 1; or (2) cyclohexyl magnesium halide addition to phenylglyoxylate as shown in Scheme 2. ##STR2##
Asymmetric synthesis of individual enantiomers of CHPGA has been approached along the lines of Scheme 2, by Grignard addition to a chiral auxiliary ester of glyoxylic acid to give a diastereomeric mixture of esters. In addition, a multiple step asymmetric synthesis of (R)-CHPGA from (D)-arabinose using Grignard reagents has been reported.
As outlined in Scheme 3 below, the simple chiral ester wherein R* is the residue of a chiral alcohol, can be directly converted to chiral drugs or drug candidates by trans-esterification (R'=acetate), or hydrolyzed to yield chiral CHPGA and then esterified (R'=H). ##STR3##
While the aforementioned asymmetric synthetic methods are adequate for many purposes, the chemical yields are in some cases poor, and the stereoselectivity is not always high. Also, the chiral auxiliary reagents that give good yields and higher stereoselectivity are often quite expensive. Thus, these processes are often cost prohibitive for use in commercial scale production of chiral pharmaceutical compounds.
A potential alternative to asymmetric synthesis is resolution of racemic CHPGA. This has been accomplished on an analytical scale using resolving agents such as ephedrine, quinine, and (+) and (-)-amphetamine. However, such resolving agents are expensive, making known processes for resolution as impractical as known asymmetric syntheses. In addition, resolution processes using these agents provide poor stereoselectivity. The poor stereoselectivity necessitates multiple recrystallization steps to isolate the single CHPGA enantiomer, which adds to the production costs of chiral pharmaceuticals made from these precursors.
A more efficient and economic method for producing .alpha.-alkylphenylglycolic acids, particularly single enantiomers of .alpha.-alkylphenylglycolic acids, on an industrial scale is therefore desirable. Such a method should provide high purity compounds in high chemical yields with few processing steps.